Mercurial > hg > documentViewer
changeset 84:a6e4f9b6729a
first version with new full-text infrastructure and slightly changed templates
| author | casties |
|---|---|
| date | Fri, 19 Mar 2010 12:42:40 +0100 |
| parents | ec12a2440daa |
| children | d3d7d05dfc54 |
| files | documentViewer.py |
| diffstat | 1 files changed, 203 insertions(+), 210 deletions(-) [+] |
line wrap: on
line diff
--- a/documentViewer.py Mon Mar 08 16:02:33 2010 +0100 +++ b/documentViewer.py Fri Mar 19 12:42:40 2010 +0100 @@ -12,6 +12,7 @@ import Ft.Xml.XPath +import cStringIO import xmlrpclib import os.path import sys @@ -22,6 +23,7 @@ import urlparse from types import * + def logger(txt,method,txt2): """logging""" logging.info(txt+ txt2) @@ -45,6 +47,14 @@ rc = rc + node.data return rc +def serializeNode(node, encoding='utf-8'): + """returns a string containing node as XML""" + buf = cStringIO.StringIO() + Print(node, stream=buf, encoding=encoding) + s = buf.getvalue() + buf.close() + return s + def getParentDir(path): """returns pathname shortened by one""" @@ -78,7 +88,9 @@ # templates and forms viewer_main = PageTemplateFile('zpt/viewer_main', globals()) thumbs_main = PageTemplateFile('zpt/thumbs_main', globals()) - image_main = PageTemplateFile('zpt/image_main', globals()) + image_main = PageTemplateFile('zpt/image_main', globals()) # obsolete! + page_main_images = PageTemplateFile('zpt/page_main_images', globals()) + page_main_text = PageTemplateFile('zpt/page_main_text', globals()) head_main = PageTemplateFile('zpt/head_main', globals()) docuviewer_css = PageTemplateFile('css/docuviewer.css', globals()) info_xml = PageTemplateFile('zpt/info_xml', globals()) @@ -88,24 +100,34 @@ changeDocumentViewerForm = PageTemplateFile('zpt/changeDocumentViewer', globals()) - def __init__(self,id,imageViewerUrl,textViewerUrl=None,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=10,authgroups="mpiwg"): + def __init__(self,id,imageScalerUrl=None,textServerName=None,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=10,authgroups="mpiwg"): """init document viewer""" self.id=id self.title=title - self.imageViewerUrl=imageViewerUrl - self.textViewerUrl=textViewerUrl - - if not digilibBaseUrl: - self.digilibBaseUrl = self.findDigilibUrl() - else: - self.digilibBaseUrl = digilibBaseUrl self.thumbcols = thumbcols self.thumbrows = thumbrows # authgroups is list of authorized groups (delimited by ,) self.authgroups = [s.strip().lower() for s in authgroups.split(',')] - # add template folder so we can always use template.something - self.manage_addFolder('template') - + # create template folder so we can always use template.something + + templateFolder = Folder('template') + #self['template'] = templateFolder # Zope-2.12 style + self._setObject('template',templateFolder) # old style + try: + from Products.XMLRpcTools.XMLRpcTools import XMLRpcServerProxy + xmlRpcClient = XMLRpcServerProxy(id='fulltextclient', serverUrl=textServerName, use_xmlrpc=False) + #templateFolder['fulltextclient'] = xmlRpcClient + templateFolder._setObject('fulltextclient',xmlRpcClient) + except Exception, e: + logging.error("Unable to create XMLRpcTools for fulltextclient: "+str(e)) + try: + from Products.zogiLib.zogiLib import zogiLib + zogilib = zogiLib(id="zogilib", title="zogilib for docuviewer", dlServerURL=imageScalerUrl, layout="book") + #templateFolder['zogilib'] = zogilib + templateFolder._setObject('zogilib',zogilib) + except Exception, e: + logging.error("Unable to create zogiLib for zogilib: "+str(e)) + security.declareProtected('View','thumbs_rss') def thumbs_rss(self,mode,url,viewMode="auto",start=None,pn=1): @@ -116,7 +138,7 @@ @param viewMode: if images display images, if text display text, default is images (text,images or auto) ''' - logging.info("HHHHHHHHHHHHHH:load the rss") + logging.debug("HHHHHHHHHHHHHH:load the rss") logger("documentViewer (index)", logging.INFO, "mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) if not hasattr(self, 'template'): @@ -139,7 +161,7 @@ return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode) security.declareProtected('View','index_html') - def index_html(self,mode,url,viewMode="auto",start=None,pn=1,mk=None): + def index_html(self,url,mode="texttool",viewMode="auto",start=None,pn=1,mk=None): ''' view it @param mode: defines how to access the document behind url @@ -148,13 +170,14 @@ ''' - logger("documentViewer (index)", logging.INFO, "mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) + logging.debug("documentViewer (index) mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) if not hasattr(self, 'template'): - # create template folder if it doesn't exist - self.manage_addFolder('template') + # this won't work + logging.error("template folder missing!") + return "ERROR: template folder missing!" - if not self.digilibBaseUrl: + if not getattr(self, 'digilibBaseUrl', None): self.digilibBaseUrl = self.findDigilibUrl() or "http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary" docinfo = self.getDocinfo(mode=mode,url=url) @@ -162,7 +185,7 @@ pt = getattr(self.template, 'viewer_main') if viewMode=="auto": # automodus gewaehlt - if docinfo.get("textURL",'') and self.textViewerUrl: #texturl gesetzt und textViewer konfiguriert + if docinfo.get("textURL",''): #texturl gesetzt und textViewer konfiguriert viewMode="text" else: viewMode="images" @@ -179,6 +202,19 @@ for m in mk: ret+="mk=%s"%m return ret + + def findDigilibUrl(self): + """try to get the digilib URL from zogilib""" + url = self.template.zogilib.getDLBaseUrl() + return url + + def getStyle(self, idx, selected, style=""): + """returns a string with the given style and append 'sel' if path == selected.""" + #logger("documentViewer (getstyle)", logging.INFO, "idx: %s selected: %s style: %s"%(idx,selected,style)) + if idx == selected: + return style + 'sel' + else: + return style def getLink(self,param=None,val=None): """link to documentviewer with parameter param set to val""" @@ -189,7 +225,8 @@ del params[param] else: params[param] = str(val) - if params["mode"] == "filepath": #wenn beim erst Aufruf filepath gesetzt wurde aendere das nun zu imagepath + + if params.get("mode", None) == "filepath": #wenn beim erst Aufruf filepath gesetzt wurde aendere das nun zu imagepath params["mode"] = "imagepath" params["url"] = getParentDir(params["url"]) @@ -225,20 +262,6 @@ return pt(docinfo=docinfo) - def getStyle(self, idx, selected, style=""): - """returns a string with the given style and append 'sel' if path == selected.""" - #logger("documentViewer (getstyle)", logging.INFO, "idx: %s selected: %s style: %s"%(idx,selected,style)) - if idx == selected: - return style + 'sel' - else: - return style - - def getTextLanguage(self,url,docinfo): - docinfo['lang']=getTextFromNode(dom.xpath("//bib/lang")[0]) - lang = urlencode({'':docinfo['lang']}) - return lang - - def isAccessible(self, docinfo): """returns if access to the resource is granted""" access = docinfo.get('accessType', None) @@ -291,6 +314,8 @@ docinfo['numPages'] = int(getTextFromNode(sizes[0])) else: docinfo['numPages'] = 0 + + # TODO: produce and keep list of image names and numbers return docinfo @@ -431,109 +456,106 @@ return docinfo - def getNumPages(self, xquery, docinfo=None): #New Method 24.02.2010 - text=self.viewerTemplates.query.eval("/mpdl/interface/xquery.xql","document="+ docinfo['textURLPath'] +"&xquery="+str(xquery)) - docinfo['numPages'] = text.count("<pb ") - return docinfo - + def getDocinfoFromTextTool(self, url, dom=None, docinfo=None): + """parse texttool tag in index meta""" + logger("documentViewer (getdocinfofromtexttool)", logging.INFO, "url: %s" % (url)) + if docinfo is None: + docinfo = {} + + if docinfo.get('lang', None) is None: + docinfo['lang'] = '' # default keine Sprache gesetzt + if dom is None: + dom = self.getIndexMeta(url) + + archivePath = None + archiveName = None + + archiveNames = dom.xpath("//resource/name") + if archiveNames and (len(archiveNames) > 0): + archiveName = getTextFromNode(archiveNames[0]) + else: + logger("documentViewer (getdocinfofromtexttool)", logging.WARNING, "resource/name missing in: %s" % (url)) + + archivePaths = dom.xpath("//resource/archive-path") + if archivePaths and (len(archivePaths) > 0): + archivePath = getTextFromNode(archivePaths[0]) + # clean up archive path + if archivePath[0] != '/': + archivePath = '/' + archivePath + if archiveName and (not archivePath.endswith(archiveName)): + archivePath += "/" + archiveName + else: + # try to get archive-path from url + logger("documentViewer (getdocinfofromtexttool)", logging.WARNING, "resource/archive-path missing in: %s" % (url)) + if (not url.startswith('http')): + archivePath = url.replace('index.meta', '') + + if archivePath is None: + # we balk without archive-path + raise IOError("Missing archive-path (for text-tool) in %s" % (url)) - def getDocinfoFromTextTool(self,url,dom=None,docinfo=None): - """parse texttool tag in index meta""" - logger("documentViewer (getdocinfofromtexttool)", logging.INFO,"url: %s"%(url)) - if docinfo is None: - docinfo = {} - - if docinfo.get('lang',None) is None: - docinfo['lang']='' # default keine Sprache gesetzt - if dom is None: - dom = self.getIndexMeta(url) - - archivePath = None - archiveName = None + imageDirs = dom.xpath("//texttool/image") + if imageDirs and (len(imageDirs) > 0): + imageDir = getTextFromNode(imageDirs[0]) + + else: + # we balk with no image tag / not necessary anymore because textmode is now standard + #raise IOError("No text-tool info in %s"%(url)) + imageDir = "" + #xquery="//pb" + docinfo['imagePath'] = "" # keine Bilder + docinfo['imageURL'] = "" + + if imageDir and archivePath: + #print "image: ", imageDir, " archivepath: ", archivePath + imageDir = os.path.join(archivePath, imageDir) + imageDir = imageDir.replace("/mpiwg/online", '') + docinfo = self.getDirinfoFromDigilib(imageDir, docinfo=docinfo) + docinfo['imagePath'] = imageDir + + docinfo['imageURL'] = self.digilibBaseUrl + "/servlet/Scaler?fn=" + imageDir + + viewerUrls = dom.xpath("//texttool/digiliburlprefix") + if viewerUrls and (len(viewerUrls) > 0): + viewerUrl = getTextFromNode(viewerUrls[0]) + docinfo['viewerURL'] = viewerUrl + + textUrls = dom.xpath("//texttool/text") + if textUrls and (len(textUrls) > 0): + textUrl = getTextFromNode(textUrls[0]) + if urlparse.urlparse(textUrl)[0] == "": #keine url + textUrl = os.path.join(archivePath, textUrl) + # fix URLs starting with /mpiwg/online + if textUrl.startswith("/mpiwg/online"): + textUrl = textUrl.replace("/mpiwg/online", '', 1) + + docinfo['textURL'] = textUrl + + textUrls = dom.xpath("//texttool/text-url-path") + if textUrls and (len(textUrls) > 0): + textUrl = getTextFromNode(textUrls[0]) + docinfo['textURLPath'] = textUrl + + presentationUrls = dom.xpath("//texttool/presentation") + docinfo = self.getBibinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get info von bib tag + + if presentationUrls and (len(presentationUrls) > 0): # ueberschreibe diese durch presentation informationen + # presentation url ergiebt sich ersetzen von index.meta in der url der fuer die Metadaten + # durch den relativen Pfad auf die presentation infos + presentationPath = getTextFromNode(presentationUrls[0]) + if url.endswith("index.meta"): + presentationUrl = url.replace('index.meta', presentationPath) + else: + presentationUrl = url + "/" + presentationPath + docinfo = self.getNumPages(docinfo) #im moment einfach auf eins setzen, navigation ueber die thumbs geht natuerlich nicht + docinfo = self.getBibinfoFromTextToolPresentation(presentationUrl, docinfo=docinfo, dom=dom) + + docinfo = self.getAuthinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get access info + + return docinfo - archiveNames=dom.xpath("//resource/name") - if archiveNames and (len(archiveNames)>0): - archiveName=getTextFromNode(archiveNames[0]) - else: - logger("documentViewer (getdocinfofromtexttool)", logging.WARNING,"resource/name missing in: %s"%(url)) - - archivePaths=dom.xpath("//resource/archive-path") - if archivePaths and (len(archivePaths)>0): - archivePath=getTextFromNode(archivePaths[0]) - # clean up archive path - if archivePath[0] != '/': - archivePath = '/' + archivePath - if archiveName and (not archivePath.endswith(archiveName)): - archivePath += "/" + archiveName - else: - # try to get archive-path from url - logger("documentViewer (getdocinfofromtexttool)", logging.WARNING,"resource/archive-path missing in: %s"%(url)) - if (not url.startswith('http')): - archivePath = url.replace('index.meta', '') - - if archivePath is None: - # we balk without archive-path - raise IOError("Missing archive-path (for text-tool) in %s"%(url)) - - imageDirs=dom.xpath("//texttool/image") - if imageDirs and (len(imageDirs)>0): - imageDir=getTextFromNode(imageDirs[0]) - - else: - # we balk with no image tag / not necessary anymore because textmode is now standard - #raise IOError("No text-tool info in %s"%(url)) - imageDir="" - #xquery="//pb" - docinfo['imagePath'] = "" # keine Bilder - docinfo['imageURL'] = "" - - if imageDir and archivePath: - #print "image: ", imageDir, " archivepath: ", archivePath - imageDir=os.path.join(archivePath,imageDir) - imageDir=imageDir.replace("/mpiwg/online",'') - docinfo=self.getDirinfoFromDigilib(imageDir,docinfo=docinfo) - docinfo['imagePath'] = imageDir - - docinfo['imageURL'] = self.digilibBaseUrl+"/servlet/Scaler?fn="+imageDir - - viewerUrls=dom.xpath("//texttool/digiliburlprefix") - if viewerUrls and (len(viewerUrls)>0): - viewerUrl=getTextFromNode(viewerUrls[0]) - docinfo['viewerURL'] = viewerUrl - - textUrls=dom.xpath("//texttool/text") - if textUrls and (len(textUrls)>0): - textUrl=getTextFromNode(textUrls[0]) - if urlparse.urlparse(textUrl)[0]=="": #keine url - textUrl=os.path.join(archivePath,textUrl) - # fix URLs starting with /mpiwg/online - if textUrl.startswith("/mpiwg/online"): - textUrl = textUrl.replace("/mpiwg/online",'',1) - - docinfo['textURL'] = textUrl - - textUrls=dom.xpath("//texttool/text-url-path") - if textUrls and (len(textUrls)>0): - textUrl=getTextFromNode(textUrls[0]) - docinfo['textURLPath'] = textUrl - - presentationUrls=dom.xpath("//texttool/presentation") - docinfo = self.getBibinfoFromIndexMeta(url,docinfo=docinfo,dom=dom) # get info von bib tag - - if presentationUrls and (len(presentationUrls)>0): # ueberschreibe diese durch presentation informationen - # presentation url ergiebt sich ersetzen von index.meta in der url der fuer die Metadaten - # durch den relativen Pfad auf die presentation infos - presentationPath = getTextFromNode(presentationUrls[0]) - if url.endswith("index.meta"): - presentationUrl=url.replace('index.meta',presentationPath) - else: - presentationUrl=url + "/" + presentationPath - docinfo=self.getNumPages('//pb', docinfo) #im moment einfach auf eins setzen, navigation ueber die thumbs geht natuerlich nicht - docinfo = self.getBibinfoFromTextToolPresentation(presentationUrl,docinfo=docinfo,dom=dom) - docinfo = self.getAuthinfoFromIndexMeta(url,docinfo=docinfo,dom=dom) # get access info - - return docinfo + def getBibinfoFromTextToolPresentation(self,url,docinfo=None,dom=None): @@ -619,7 +641,6 @@ pageinfo['start'] = start pageinfo['end'] = start + grpsize if docinfo is not None: - np = int(docinfo['numPages']) pageinfo['end'] = min(pageinfo['end'], np) pageinfo['numgroups'] = int(np / grpsize) @@ -628,82 +649,57 @@ return pageinfo - def text(self,mode,url,pn): - """give text""" - if mode=="texttool": #index.meta with texttool information - (viewerUrl,imagepath,textpath)=parseUrlTextTool(url) - - #print textpath - try: - dom = NonvalidatingReader.parseUri(textpath) - except: - return None - - list=[] - nodes=dom.xpath("//pb") + - node=nodes[int(pn)-1] - - p=node - - while p.tagName!="p": - p=p.parentNode - - - endNode=nodes[int(pn)] - - - e=endNode - - while e.tagName!="p": - e=e.parentNode - - - next=node.parentNode + def getNumPages(self,docinfo=None): + """get list of pages from fulltext and put in docinfo""" + xquery = '//pb' + text = self.template.fulltextclient.eval("/mpdl/interface/xquery.xql", "document=%s&xquery=%s"%(docinfo['textURLPath'],xquery)) + # TODO: better processing of the page list. do we need the info somewhere else also? + docinfo['numPages'] = text.count("<pb ") + return docinfo + + def getTextPage(self, mode="text", pn=1, docinfo=None): + """returns single page from fulltext""" + pagexml=self.template.fulltextclient.eval("/mpdl/interface/page-fragment.xql", "document=%s&mode=%s&pn=%s"%(docinfo['textURLPath'],mode,pn), outputUnicode=False) + # post-processing downloaded xml + pagedom = Parse(pagexml) + # plain text mode + if mode == "text": + # first div contains text + pagedivs = pagedom.xpath("/div") + if len(pagedivs) > 0: + pagenode = pagedivs[0] + return serializeNode(pagenode) + + # text-with-links mode + if mode == "textPollux": + # first div contains text + pagedivs = pagedom.xpath("/div") + if len(pagedivs) > 0: + pagenode = pagedivs[0] + # check all a-tags + links = pagenode.xpath("//a") + for l in links: + hrefNode = l.getAttributeNodeNS(None, u"href") + if hrefNode: + # is link with href + href = hrefNode.nodeValue + if href.startswith('lt/lex.xql'): + # is pollux link + selfurl = self.absolute_url() + # change href + hrefNode.nodeValue = href.replace('lt/lex.xql','%s/head_main_voc'%selfurl) + # add target + l.setAttributeNS(None, 'target', '_blank') + return serializeNode(pagenode) - #sammle s - while next and (next!=endNode.parentNode): - list.append(next) - next=next.nextSibling - list.append(endNode.parentNode) - - if p==e:# beide im selben paragraphen - pass -# else: -# next=p -# while next!=e: -# print next,e -# list.append(next) -# next=next.nextSibling -# -# for x in list: -# PrettyPrint(x) -# -# return list -# + return "no text here" - def findDigilibUrl(self): - """try to get the digilib URL from zogilib""" - url = self.imageViewerUrl[:-1] + "/getScalerUrl" - #print urlparse.urlparse(url)[0] - #print urlparse.urljoin(self.absolute_url(),url) - logging.info("finddigiliburl: %s"%urlparse.urlparse(url)[0]) - logging.info("finddigiliburl: %s"%urlparse.urljoin(self.absolute_url(),url)) - - try: - if urlparse.urlparse(url)[0]=='': #relative path - url=urlparse.urljoin(self.absolute_url()+"/",url) - - scaler = urlopen(url).read() - return scaler.replace("/servlet/Scaler?", "") - except: - return None - def changeDocumentViewer(self,imageViewerUrl,textViewerUrl,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=10,authgroups='mpiwg',RESPONSE=None): + def changeDocumentViewer(self,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=10,authgroups='mpiwg',RESPONSE=None): """init document viewer""" self.title=title - self.imageViewerUrl=imageViewerUrl - self.textViewerUrl=textViewerUrl self.digilibBaseUrl = digilibBaseUrl self.thumbrows = thumbrows self.thumbcols = thumbcols @@ -713,17 +709,14 @@ - -# security.declareProtected('View management screens','renameImageForm') - def manage_AddDocumentViewerForm(self): """add the viewer form""" pt=PageTemplateFile('zpt/addDocumentViewer', globals()).__of__(self) return pt() -def manage_AddDocumentViewer(self,id,imageViewerUrl="",textViewerUrl="",title="",RESPONSE=None): +def manage_AddDocumentViewer(self,id,imageScalerUrl="",textServerName="",title="",RESPONSE=None): """add the viewer""" - newObj=documentViewer(id,imageViewerUrl,title=title,textViewerUrl=textViewerUrl) + newObj=documentViewer(id,imageScalerUrl=imageScalerUrl,title=title,textServerName=textServerName) self._setObject(id,newObj) if RESPONSE is not None: